The heavily instrumented full scale caisson (9 m high), constructed at the port of Kushiro (Kushiro, Hokkaido, Japan) in 2001, was impacted by a major earthquake hitting the site in September 2003 (M = 7.5 with the maximum acceleration of 143.5 gals) where the dynamic water pressures from both backfill and seawater side were recorded. While comparing the recorded dynamic water pressure with Westergaad’s theoretical solution [1], noticeable discrepancy was observed between compression phases of applied dynamic pressure on wall whereas an excellent correspondence was noticed in dynamic suction. In this paper, the soil-seawater-caisson system has also been numerically modeled using Finite Element Technique. The recorded data in terms of applied ocean water pressure on wall is analyzed and compared with numerical results as well as Westergard solution. The results demonstrate that, the recorded dynamic ocean water pressure was rather close to Westergard’s solution in suction phase while in the compression phase the former is much higher. In addition, according to numerical results, dynamic ocean water pressure is over-predicted either in compression or suction phase, however, the level of over-perdition is higher in compression phase than suction phase. Furthermore, the effect of vertical component of earthquake is investigated. The results show that the addition of the vertical component of earthquake increases the compression dynamic pressure while the suction pressure is not affected.

This content is only available via PDF.
You do not currently have access to this content.